Towards Understanding the EEG-fMRI Relationship Through Biologically Grounded Whole-Brain Modeling
Poster No:
1340
Submission Type:
Abstract Submission
Authors:
Stanislav Jiricek1,2,3, Vincent Chien1, Helmut Schmidt1, Vlastimil Koudelka2, Radek Marecek4, Stella Sánchez1, Dante Mantini5, Jaroslav Hlinka2,1
Institutions:
1Institute of Computer Science of the Czech Academy of Sciences, Prague, Czech Republic, 2National Institute of Mental Health, Klecany, Czech Republic, 3Faculty of Electrical Engineering, Czech Technical University in Prague, Prague, Czech Republic, 4Central European Institute of Technology (CEITEC), Masaryk University, Brno, Czech Republic, 5Movement Control and Neuroplasticity Research Group, KU Leuven, Leuven, Belgium
First Author:
Stanislav Jiricek
Institute of Computer Science of the Czech Academy of Sciences|National Institute of Mental Health|Faculty of Electrical Engineering, Czech Technical University in Prague
Prague, Czech Republic|Klecany, Czech Republic|Prague, Czech Republic
Institute of Computer Science of the Czech Academy of Sciences|National Institute of Mental Health|Faculty of Electrical Engineering, Czech Technical University in Prague
Prague, Czech Republic|Klecany, Czech Republic|Prague, Czech Republic
Co-Author(s):
Helmut Schmidt
Institute of Computer Science of the Czech Academy of Sciences
Prague, Czech Republic
Institute of Computer Science of the Czech Academy of Sciences
Prague, Czech Republic
Radek Marecek
Central European Institute of Technology (CEITEC), Masaryk University
Brno, Czech Republic
Central European Institute of Technology (CEITEC), Masaryk University
Brno, Czech Republic
Stella Sánchez
Institute of Computer Science of the Czech Academy of Sciences
Prague, Czech Republic
Institute of Computer Science of the Czech Academy of Sciences
Prague, Czech Republic
Jaroslav Hlinka
National Institute of Mental Health|Institute of Computer Science of the Czech Academy of Sciences
Klecany, Czech Republic|Prague, Czech Republic
National Institute of Mental Health|Institute of Computer Science of the Czech Academy of Sciences
Klecany, Czech Republic|Prague, Czech Republic
Introduction:
The link between concurrently measured resting-state EEG and fMRI signals remains elusive. Only a single consistent EEG-BOLD correlation pattern has emerged: anticorrelation between parieto-occipital EEG alpha power and BOLD fluctuations, primarily in sensory areas (Laufs et al., 2003; Jiricek et al., 2024). Recent computational models aimed to uncover the mechanisms of EEG-BOLD (anti)correlations. Proposed mechanisms vary across modeling frameworks (Schirner et al., 2018; Pang et al., 2018; Rabuffo et al., 2021). We developed a whole-brain network model with sufficient detail to simulate EEG and fMRI data, incorporating cortical heterogeneity. Specifically, we performed:
1) Parameter optimization: Fit the model to empirical EEG spectra and fMRI BOLD functional connectivity (FC).
2) EEG-BOLD pattern replication: Test the model's ability to replicate empirical EEG alpha-BOLD relationship.
3) Sensitivity and knock-out analysis: Identification of critical model elements underlying simulated EEG, BOLD, and their correlation.
1) Parameter optimization: Fit the model to empirical EEG spectra and fMRI BOLD functional connectivity (FC).
2) EEG-BOLD pattern replication: Test the model's ability to replicate empirical EEG alpha-BOLD relationship.
3) Sensitivity and knock-out analysis: Identification of critical model elements underlying simulated EEG, BOLD, and their correlation.
Methods:
For model definition, optimization, and evaluation, we derived features from multiple data modalities across four studies, interpolated to AAL90 atlas centroids, focusing on left hemisphere cortical areas: 1) EEG-fMRI dataset: EEG spectra and BOLD FC matrices for optimization, source-localized EEG alpha-BOLD correlation maps for evaluation (Jiricek et al., 2024); 2) Structural Connectivity (SC) dataset: Average SC matrix from processed DWI data of 88 subjects (Škoch et al., 2022); 3) Cortical column connectivity dataset: 8x8 connectivity matrix for neuronal populations (PYR, PV, SST, VIP) in layers L2/3 and L5 (Jiang et al., 2015; Hahn et al., 2022); 4) MRI-derived T1w/T2w dataset: Single T1/T2 map for cortical heterogeneities (Burt et al., 2018), see Figure 1. Each brain area's dynamics were defined by a rate-based model incorporating local microcircuit dynamics (I_loc), input from other areas via the SC matrix and T1/T2-defined feedforward and feedback connections, and constant T1/T2-scaled thalamic input. Independent Gaussian noise for each area and population was used to drive the model.
·Fig. 1: Visualization of derived features based on different modalities and datasets.
Results:
We optimized 28 parameters using MATLAB's genetic algorithm, based on Spearman correlation fit to EEG spectra (occipital regions) and BOLD FC. The achieved fits were R_EEG = 0.99 for EEG spectra and R_BOLD = 0.64 for BOLD FC (Fig. 2B). The simulated EEG-BOLD correlation pattern matched empirical data (median R = 0.16, p < 0.001), validating the model (Fig. 2B). The simulated pattern correlated with the T1/T2 map, linking cortical hierarchy to EEG alpha-BOLD correlations. Local model heterogeneities also contributed to optimization (Fig. 1).
Sensitivity analysis of EEG alpha-BOLD correlation (first-order Sobol indices) identified key parameters: the slope of the input-output rate function, SST and PYR time constants, PV feedforward weights, delay matrix scaling, and SST population heterogeneity (Fig. 2C). Knock-out analysis revealed critical local connections, including PYR loops with SST and PV populations, and key inter-layer connections (Fig. 2D). Four essential inter-area connections targeting superficial layer PV and SST, and deep-layer PYR and SST populations were also identified (Fig. 2E).
Sensitivity analysis of EEG alpha-BOLD correlation (first-order Sobol indices) identified key parameters: the slope of the input-output rate function, SST and PYR time constants, PV feedforward weights, delay matrix scaling, and SST population heterogeneity (Fig. 2C). Knock-out analysis revealed critical local connections, including PYR loops with SST and PV populations, and key inter-layer connections (Fig. 2D). Four essential inter-area connections targeting superficial layer PV and SST, and deep-layer PYR and SST populations were also identified (Fig. 2E).
·Fig. 2: Model's (A) alpha-BOLD corr. (B) Emp. vs sim. spectra, and FC. Sim. vs emp. alpha-BOLD corr. (C) Sensitivity for alpha-BOLD corr. (D) Local, (E) inter-area conn. spectra knock-out analysis.
Conclusions:
We proposed a novel whole-brain network computational model for multimodal EEG-fMRI data. The model demonstrates a high degree of fit to the typical features of EEG and fMRI. It replicated the empirical EEG alpha-BOLD correlation, supporting its validity. Sensitivity and knock-out analyses identified key parameters for further investigation. The model enables the study of mechanisms underlying the EEG-BOLD relationship and can be applied to other EEG-fMRI datasets to provide mechanistic insights into various conditions.
Acknowledgements
The study was supported by ERDF-Project Brain dynamics, No. CZ.02.01.01/00/22_008/0004643, and the Czech Technical
University Internal Grant Agency - grant number SGS23/120/OHK3/2T/13.
Acknowledgements
The study was supported by ERDF-Project Brain dynamics, No. CZ.02.01.01/00/22_008/0004643, and the Czech Technical
University Internal Grant Agency - grant number SGS23/120/OHK3/2T/13.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
EEG/MEG Modeling and Analysis 1
fMRI Connectivity and Network Modeling 2
Methods Development
Task-Independent and Resting-State Analysis
Keywords:
Computational Neuroscience
Cortex
Cortical Layers
Data analysis
Electroencephaolography (EEG)
FUNCTIONAL MRI
Source Localization
Other - Whole-brain network modeling
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Jiricek, S., Koudelka, V., Mantini, D., Marecek, R., & Hlinka, J. (2024). Spatial (Mis) match Between EEG and fMRI Signal Patterns Revealed by Spatio-Spectral Source-Space EEG Decomposition. bioRxiv, 2024-07.
Schirner, M., McIntosh, A. R., Jirsa, V., Deco, G., & Ritter, P. (2018). Inferring multi-scale neural mechanisms with brain network modelling. elife, 7, e28927.
Pang, J. C., & Robinson, P. A. (2018). Neural mechanisms of the EEG alpha-BOLD anticorrelation. Neuroimage, 181, 461-470.
Rabuffo, G., Fousek, J., Bernard, C., & Jirsa, V. (2021). Neuronal cascades shape whole-brain functional dynamics at rest. Eneuro, 8(5).
Škoch, A., Rehák Bučková, B., Mareš, J., Tintěra, J., Sanda, P., Jajcay, L., ... & Hlinka, J. (2022). Human brain structural connectivity matrices–ready for modelling. Scientific Data, 9(1), 486.
Hahn, G., Kumar, A., Schmidt, H., Knösche, T. R., & Deco, G. (2022). Rate and oscillatory switching dynamics of a multilayer visual microcircuit model. Elife, 11, e77594.
Jiang, X., Shen, S., Cadwell, C. R., Berens, P., Sinz, F., Ecker, A. S., ... & Tolias, A. S. (2015). Principles of connectivity among morphologically defined cell types in adult neocortex. Science, 350(6264), aac9462.
Burt, J. B., Demirtaş, M., Eckner, W. J., Navejar, N. M., Ji, J. L., Martin, W. J., ... & Murray, J. D. (2018). Hierarchy of transcriptomic specialization across human cortex captured by structural neuroimaging topography. Nature neuroscience, 21(9), 1251-1259.